Fault diagnosis in induction motors using self-organizing neural networks and quantization error

In order to effectively solve the problems such as fault with single features, carrying the lack of information resulting in poor effect of fault judgment in traditional transformer fault diagnosis method. In view of this, this paper will consider the insulation resistance of the core 12 fault symptoms and fault characteristic gases as the combination of fault characteristics, effectively solve the problem of the existence of the feature quantity; At the same time, the principal component analysis and self-organizing neural network combination is introduced to this paper, this paper puts forward a new method of transformer fault diagnosis based on principal component analysis and self-organizing neural network. Firstly, using the principal component analysis for the main components of 17 kinds of feature extraction, extracted features can effectively present the characteristics of 10 kinds of fault classification, on this basis, the 10 methods of feature extraction was obtained as input vector of self-organizing neural network and identification of fault types. The example analysis shows that this method has a good diagnostic effect, which can effectively improve the accuracy of fault diagnosis and verify the validity and feasibility of this method.

Nowadays, many industrial applications require online systems that combine several processing techniques in order to offer solutions to complex problems as the case of detection and classification of multiple faults in induction motors. In this work, a novel digital structure to implement the empirical mode decomposition (EMD) for processing nonstationary and nonlinear signals using the full spline-cubic function is presented; besides, it is combined with an adaptive linear network (ADALINE)-based frequency estimator and a feed forward neural network (FFNN)-based classifier to provide an intelligent methodology for the automatic diagnosis during the startup transient of motor faults such as: one and two broken rotor bars, bearing defects, and unbalance. Moreover, the overall methodology implementation into a field-programmable gate array (FPGA) allows an online and real-time operation, thanks to its parallelism and high-performance capabilities as a system-on-a-chip (SoC) solution. The detection and classification results show the effectiveness of the proposed fused techniques; besides, the high precision and minimum resource usage of the developed digital structures make them a suitable and low-cost solution for this and many other industrial applications.

Health monitoring of induction motors through embedded systems-simulation of broker rotor bar fault and abnormal gear teeth fault

Motors have become critical components for many industries. A motor failure can result in the shutdown of a production line, or require that redundant equipment be utilised to circumvent the problem. By this way, the early detection of motor deterioration can increase plant availability and safety in an economical way. Many publications have investigated the detection of rotor and stator winding failures in electrical machines; however, there is more limited research on fault diagnosis in induction motors fed by PWM converters. This paper proposes a method, based on spectral estimation to diagnosis stator winding faults in induction motors fed by PWM converters. Simulation and experimental results are employed to validate the method. Further, a 5 hp and 3 hp, 230V induction motors with internal tapings from the stator windings are specially constructed. The motors are powered by a PWM converter and different fault conditions are investigated.

Induction motors are workhorses of industry. Hence, fault diagnosis of an induction motor is vital in every industry for reduction of maintenance cost, safety, increase of production etc. Motor Current Signature Analysis is a non invasive online monitoring technique for the fault diagnosis of induction motors. In this paper, an Induction motor has been modeled using coupled circuit modeling of induction motor. This model is successfully used to simulate the induction motors at healthy and faulty cases. This model gives the complete information of stator current required for MCSA. Rotor broken bars results rotor asymmetry, which induces a frequency components around fundamental in stator current. Neural network is one of the reliable soft computing techniques. Based on the rotor harmonics and slip, a neural network method to diagnose the rotor broken faults of an induction motor is presented.

The motor is the workhorse of industries. The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis, and prognosis are of increasing importance. This paper introduces a technique to detect faults in induction motors. Stator currents are measured by current meters and stored by time domain. The time domain is not suitable for representing current signals, so the frequency domain is used to display the signals. Fourier transform is used to convert the signals onto frequency domain. After the signals have been converted, the features of the signals are extracted by the signal processing methods like the wavelet analysis, spectrum analysis, and other methods. The discovered features are entered to a pattern classification model such as a neural network model, a polynomial neural network, a fuzzy inference model, or other models. This paper describes the results of detecting fault using Fourier and wavelet analysis.KeywordsFault DetectionWavelet AnalysisFault DiagnosisInduction MotorWavelet DecompositionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This paper presents recent developments in the use of a new technique for the diagnosis of faults in three-phase induction motors. The proposed diagnostic technique, which is based on the multiple reference frames theory (MRFT), has been successfully used in the diagnosis of stator faults, rotor faults and air gap eccentricity in line-connected and direct torque controlled (DTC) induction motors. The use of the MRFT represents a step forward towards the unification of the diagnostic process of faults in induction motors and drives by using the same technique to diagnose any type of faults. Due to the modular structure of the MRFT, virtually it can be used to diagnose any motor fault that has a direct impact in one or more supply currents. In addition, as it was demonstrated in a previous work, the technique is particularly adequate for an easy integration into the control system of a DTC induction motor drive at no additional cost.

Hilbert transformation is an ideal phase shifting tool in data signal processing. Being Hilbert transformed, the conjugated one of a signal is obtained. The Hilbert modulus is defined as the square of a signal and its conjugation. This work presents a method by which rotor faults of squirrel cage induction motors, such as broken rotor bars and eccentricity, can be diagnosed. The method is based on the spectral analysis of the stator current Hilbert Modulus of the induction motors. Theoretical analysis and experimental results demonstrate that has the same rotor fault detecting ability as the extended Park' vector approach. The vital advantage of the former is the smaller hardware and software spending compared with the existing ones.

The induction motors found in industrial and commercial applications are responsible for most of the energy consumption in the world. These machines are widely used because of their advantages like high efficiency, robustness, and practicality; nevertheless, the occurrence of unexpected faults may affect their proper operation leading to unnecessary breakdowns with economic repercussions. For that reason, the development of methodologies that ensure their proper operation is very important, and in this sense, this paper presents an evaluation of signal entropy as an alternative fault-related feature for detecting faults in induction motors and their kinematic chain. The novelty and contribution lie in calculating a set of entropy-related features from vibration and stator current signals measured from an induction motor operating under different fault conditions. The aim of this work is to identify changes and trends in entropy-related features produced by faulty conditions such as broken rotor bars, damage in bearings, misalignment, unbalance, as well as different severities of uniform wear in gearboxes. The estimated entropy-related features are compared to other classical features in order to determine the sensitivity and potentiality of entropy in providing valuable information that could be useful in future work for developing a complete methodology for identifying and classifying faults. The performed analysis is applied to real experimental data acquired from a laboratory test bench and the obtained results depict that entropy-related features can provide significant information related to particular faults in induction motors and their kinematic chain.

Traditional Induction Motor (IM) Fault Diagnosis (FD) relies on extracting features from original signals, which directly impacts FD performance. However, high-quality features require expert knowledge and human interaction. Therefore, this paper presents a comparative analysis of the detection of stator inter-turn faults in IMs using state-of-art classifiers. Experimental data was acquired and the class was divided, into up to three levels of severity for the stator fault. Thereafter, using a non-parametric approach, features from the three-phase currents were engineered relative to the characteristic fault frequencies for stator faults. Various families of classifiers were trained on the developed feature-set, and for each family of classifiers, Bayesian Optimization was applied to train its own variants and derive the best hyperparameters. The best results were obtained for Tree, Ensembled and Shallow Neural-based techniques using a separate test set.

This paper demonstrates the development of a flexible fault diagnosis methodology that can detect up to ten different faults in the induction motor (IM), simultaneously. The major IM electrical faults, such as the broken rotor bar (BRB), phase unbalance (PUF), and stator winding fault (SWF), and mechanical faults, such as bearing fault (BF), unbalanced rotor (UR), bowed rotor (BR), and misaligned rotor (MR), are considered with different fault severities for the diagnosis. The experiments are conducted with three varying loads and seven different speeds, and the frequency domain vibration and current data are acquired at a relatively low sampling rate of 1 kHz. Several statistical features are extracted and then the best feature-set is selected using the wrapper model. Thereafter, a data classification tool based on the support vector machine (SVM) is used for the fault characterization. Initially, a multi-fault diagnosis is performed by training and testing the SVM at the same operating conditions (i.e., load and speed). The performance of the classifier is found to be very good at all IM operating conditions. The main focus of this study lies in overcoming the fault diagnosis, where the data are unavailable at required operating conditions. This is accomplished by employing interpolation and extrapolation strategies for different loads and speeds. The proposed methodology not only solves practical problem of unavailability of data at different operating conditions but also shows good performance and takes low computation time, which are vital requirements of an online intelligent condition monitoring system.

A critical analysis on broken rotor bar fault in three-phase induction motor is presented in this manuscript. Core research is carried on model-based broken bars fault detection and diagnosis for induction motors. A wide variety of faulty induction motor models are available in the literature. Out of which the model induction motor models based reference frames are widely used in fault diagnosis. It is believed to be a novel contribution, as researchers haven't considered whether these faulty models behave similar behaviour in stationary, rotor, synchronous and arbitrary reference frames. As an extension, a newly modified broken rotor induction motor model is proposed in this manuscript. To validate proposed idea dynamic computations of models are carried on using MATLAB R2022a programming software. The novel idea is verified with extensive simulation studies with various induction motors. The results and observations are presented clearly in this manuscript.

Novel indices for broken rotor bars fault diagnosis in induction motors using wavelet transform

Inadequate market segmentation and clustering problems could cause an enterprise to either miss a strategic marketing opportunity or not cash-in on a tactical campaign. The need for in-depth knowledge of customer segments and the need to overcome the limitations of using linear techniques to analyse non-linear problems requires a re-assessment of generally used approaches. The objectives of the research are (1) to consider the use of self-organising (SOM) neural networks for segmenting customer markets and (2) to analyse the predictive ability of backpropagation (BP) neural networks for classifying new customers by using the output provided by SOM neural networks. The nature and scope of neural networks are considered and the conceptual differences between Cluster Analysis and SOM neural networks as well as BP neural networks and multiple linear regression (MLR) static filter model are highlighted. The findings of the SOM neural network modelling indicate four natural clusters. In addition, the predictive ability of the BP neural network model was superior to that of MLR. Additional knowledge was also extracted from the BP neural network model by analysing the relationship between the input variables and each segment by means of input-output analysis. Sensitivity analysis was also used to identify important variables for each segment. Input-output analysis is also used to compile broad profiles of differences between the segments. The BP neural network model developed for this application is also suitable for deployment (i.e. classification of “new” customers) with a high level of confidence.

Broken rotor bar (BRB) faults are one of the most common faults in induction motors (IM). One or more broken bars can reduce the performance and efficiency of the IM and hence waste the electrical power and decrease the reliability of the whole mechanical system. This paper proposes an effective fault diagnosis method using the Teager–Kaiser energy operator (TKEO) for BRB faults detection based on the motor current signal analysis (MCSA). The TKEO is investigated and applied to remove the main supply component of the motor current for accurate fault feature extraction, especially for an IM operating at low load with low slip. Through sensing the estimation of the instantaneous amplitude (IA) and instantaneous frequency (IF) of the motor current signal using TKEO, the fault characteristic frequencies can be enhanced and extracted for the accurate detection of BRB fault severities under different operating conditions. The proposed method has been validated by simulation and experimental studies that tested the IMs with different BRB fault severities to consider the effectiveness of the proposed method. The obtained results are compared with those obtained using the conventional envelope analysis methods and showed that the proposed method provides more accurate fault diagnosis results and can distinguish the BRB fault types and severities effectively, especially for operating conditions with low loads.